^{1}

^{*}

^{1}

Ghana has been confronted with series of economic problems to the extent of calling on the IMF for a bailout after every eight years. This situation has persisted in spite of various monetary authority stabilization policies. This paper therefore focuses on investigating the games of monetary policy, inflation and economic growth of the Ghanaian economy for the period of 1982-2017. Using Autoregressive Distributed Lag (ARDL) to cointegration model, it was revealed from the study that in the long run interest rate significantly influences economic growth but in a negative direction, implies that a higher interest rate has the tendency to restrained economic growth and inflationary pressures. In relation to exchange rate, the long run result indicates an insignificant negative effect on economic growth. The general results suggest that macroeconomic variable which influences economic growth is interest rate and exchange rate. This is evidence that macroeconomic instabilities have significant effect on economic growth. This therefore calls for fiscal discipline and autonomy power to the Bank of Ghana with less interreference from the government to enable the smooth implementation of monetary policies without any string of politics attached.

Stabilization of a nation’s economy is the priority of every government. In any case, the soundness of every economy revolves around its economic and financial performance. Monetary policy alludes to the procedure by which the monetary experts of a country pedal the supply of money, regularly focusing on a rate of interest for the purpose of encouraging economic growth and stability. Countries depend heavily on their monetary policies as it serves as a key driver of economic growth through its impact on economic variables. Monetary development is crucial in any economy as it lessens poverty, which enhances the livelihoods of households. Researchers from different purviews have widely explored the connection between macroeconomic factors and their impacts on economic growth. The attractive ingredient luring these various investigations is the instability of economic activities, different policies and different regulations that influence economic growth and economic development. The central bank monetary policies influence every single business activity in an economy hence the standard of living of citizens is determined by monetary policies. Folawewo and Osinubi, [

According to Precious and Palesa [

Researchers over the years have examined the efficiency and stabilization role of monetary policy as an imperative tool for achieving a desirable macroeconomic position in both developing and developed countries. The scholars have agreed to the fact that monetary policy can have a significant impact on the growth of the economy through price stability. However, these studies including Tobin [

It is amazing up to date the convergence between monetary policy and various macroeconomic variables still differ between policymakers and economist withstanding the essential roles that monetary tools play to warrant effective regulations of the economy (Owolabi & Adegbite) [

Bawumia and Abradu-Otoo [

Starr [

Dele [

Rafiq and Mallick [

Fasanya, Onakoya and Agboluaje [

Chiaraah and Nkegbe [

Nchor, Darkwah and Lubo Sstrelec [

Awan and Asalam [

Most recently, Ayodeji and Oluwole [

It is obvious from literature that as previous studies have pursued to effectively analyze in terms of variable measurements and choice of model, there are still gaps. This study therefore is further into the roles of inflation, exchange rate, money supply and interest rate on economic growth of Ghana.

This paper makes use of time series data from the period of 1982 to 2017. Data has been gathered from World Bank and Bank of Ghana annual report. In order to determine the game played by monetary policy and inflation on economic growth of the Ghanaian economy, this paper employed and modify the model in the work of Ayodeji and Oluwole [

G D P = ƒ ( I N F , E X R , M S , I R , e ) (1)

where, G D P denotes the Gross Domestic Product, I N F represent inflation which is Ghana’s general consumer price levels, E X R denotes exchange rate, M S represents money supply which is broad money (M2); I R as interest rate representing the monetary policy rate and e as the stochastic error term that captures all relevant potential variables that were omitted from the model. The model is transformed into an econometric model as;

G D P = β 0 + β 1 I N F + β 2 E X R + β 3 M S + β 4 I R + ε t (2)

β 0 is the constant intercept and β 1 , β 2 , β 3 a n d β 4 are the parameter elasticity coefficients. Again, to enable the interpretation of the partial elasticities, which addresses the degree of responsiveness of the dependent variable to the respective economic independent variable, it is essential to introduce log into the model. This will transform the model into;

ln G D P = β 0 + β 1 ln I N F + β 2 ln E X R + β 3 ln M S + β 4 ln I R + ε t (3)

Gross Domestic Product (GDP) there are several ways to determine the growth pace of an economy. Gross domestic product is one of the basic methods to used and most preferred as its influence attributes to one’s ability to command goods and services. This paper adopts GDP as the proxy to measure the economic growth and used as the dependent variable.

Inflation (INF) in this paper is an independent variable and it’s measured by consumer price index. With high inflation the cost of goods and services increase thereby leading to an increase in hardship for the consumers. This paper expects inflation to be positive on economic growth of Ghana.

Exchange Rate (EXR) the appreciation of domestic currency in the international market enables export to be expensive and hence leading a low demand for exports and on the other hand imports become cheaper hence discouraging the desire for domestic products. This paper anticipates both positive and negative impact on the economic growth.

Money Supply (MS) according to the simple theoretical monetary model an increase in money supply (broad money) reflects an expansionary monetary policy that leads to an increase in output but associated with inflationary effect (Ofori-Abebrese et al., [

Interest Rate (IR) is also an independent variable which represents the monetary policy rate of Ghana. A lower interest rate creates a ripple effect of increase spending in an economy. Citizens will be willing to purchase and borrow more with a low interest rate and opposite when interest rate is high. The sign of IR can either be positive or negative.

In this paper, Augmented Dickey-Fuller (ADF) was employed to warrant the consistency of exploring the study by evaluating the stationary properties of the variables mentioned above in order to avoid biased, false and misleading results. In order to examine the long-run relationship between gross domestic product and the independent variables, the ARDL bound testing approach developed by Pesaran and Shin [

ln G D P t = ω 0 + δ 1 ln I N F t − 1 + δ 2 ln E X R t − 1 + δ 3 ln M S t − 1 + δ 4 ln I R t − 1 + ∑ i = 1 P ω 1 i ln I N F t − i + ∑ j = 1 q ω 2 j ln E X R t − j + ∑ k = 1 q ω 3 k ln M S t − k + ∑ l = 1 q ω 4 l ln I R t − l + μ t (4)

where ω 0 is the difference of the exogenous variables are the short run multipliers/dynamics of the model to be estimated through ECM and δ i represents the long run multipliers. The term ω 0 is the constant and the μ t denotes the error term.

The ARDL bound test consists of three stages. The initial stage is the testing for the presence of long run correlation among the variables by estimating Equation (4) by the use of OLS. The F-Test is conducted for couple significance of the numerical values of the lagged levels of the variables to enable the testing for the presence of long rung correlation among variables. The following hypothesis has been deduced;

H 0 : δ 1 = δ 2 = δ 3 = δ 4 = 0 ≫ Not cointegrated

H 1 : δ 1 = δ 2 = δ 3 = δ 4 ≠ 0

A test for cointegration is provided by two asymptotic critical values where the exogenous variables are I(m) (where 0 ≤ m ≤ 1 ). The explanatory variables are assumed to be integrated of order zero, I(0) by the lower bound values and integrated of order I(1) by the upper bound values. The null hypothesis of no cointegration will be rejected if the F-statistics is above the upper bound and accepted if it falls below the lower bound.

The second step is to test for the long run correlation after establishing the presence of cointegration. The conditional version of the ARDL model order ( m , n 1 , n 2 , n 3 ) is presented below;

G D P t = ω 0 + ∑ δ 1 ln I N F t − 1 + ∑ δ 2 ln E X R t − 1 + ∑ δ 3 ln M S t − 1 + ∑ δ 4 ln I R t − 1 + μ t (4)

The lag length of the variables was selected based on the Schwarz Bayesian criterion.

The short run dynamics is employed by the error correction model. This is the final stage of the ARDL model estimation. The model is as follows;

Δ G D P t = ω 0 + ∑ ω 1 i Δ ln I N F t − 1 + ∑ ω 2 i Δ ln E X R t − 1 + ∑ ω 3 i Δ ln M S t − 1 + ∑ ω 4 i Δ ln I R t − 1 + θ E C M t − i + μ t (5)

where ω i is the short run coefficient of model’s dynamic adjustment to equilibrium. E C M t − 1 term is Error Correction factor. It portrays the estimate of short run disequilibrium adjustment of the long run equilibrium error term. θ measures speed of adjustment to ascertain equilibrium is the problem of shock.

This section of the paper consists of analyzed estimation of different tests. Figures 1-4 display the trends of the variables.

LGDP | LBROAD_MONEY | LEXCHANGE_RATE | LINFLATION | LINTEREST_RATE | |
---|---|---|---|---|---|

Mean | 1.649184 | 3.167099 | −0.319139 | 2.924861 | 3.129608 |

Median | 1.578926 | 3.280338 | −0.164252 | 2.821219 | 3.112527 |

Maximum | 2.639057 | 3.540959 | 1.481605 | 4.085976 | 3.806662 |

Minimum | 1.193922 | 2.468100 | −2.302585 | 2.163323 | 2.525729 |

Std. Dev. | 0.324294 | 0.331299 | 0.970851 | 0.518322 | 0.361081 |

Skewness | 1.062868 | −0.671366 | 0.061195 | 0.471036 | 0.196390 |

Kurtosis | 3.950151 | 2.099101 | 2.343729 | 2.245698 | 2.312117 |

Jarque-Bera | 7.680515 | 3.703943 | 0.631367 | 2.063336 | 0.888901 |

Probability | 0.021488 | 0.156927 | 0.729290 | 0.356412 | 0.641176 |

Sum | 56.07225 | 107.6814 | −10.85074 | 99.44529 | 106.4067 |

Sum Sq. Dev. | 3.470488 | 3.622056 | 31.10420 | 8.865699 | 4.302529 |

Observations | 34 | 34 | 34 | 34 | 34 |

Source: By Authors, 2019.

VARIABLES | AUGMENTED DICKEY-FULLER TEST | LEVEL OF INTEGRATION | |
---|---|---|---|

NO TREND | TREND | ||

LEVEL | |||

BROAD_MONEY | −0.888704 | −2.646883 | - |

EXCHANGE_RATE | 1.773056 | 0.082942 | - |

GDP | −4.852887*** | −4.980694*** | I(0) |

INFLATION | −4.478889*** | −5.947897*** | I(0) |

INTEREST_RATE | −1.924857 | −2.066799 | - |

VARIABLES | AUGMENTED DICKEY-FULLER TEST | LEVEL OF INTEGRATION | |
---|---|---|---|

NO TREND | TREND | ||

FIRST DIFFERENCE | |||

BROAD_MONEY | −7.056867*** | −6.924801*** | I(1) |

EXCHANGE_RATE | −5.629902*** | −7.285127*** | I(1) |

INTEREST_RATE | −5.792823*** | −5.827045*** | I(1) |

Source: By Authors, 2019. ***at 1 percent level of significance **at 5 percent level of significance *at 10 percent level of significance.

F-Statistics | Significance | Lower Bound | Upper Bound | Decision |
---|---|---|---|---|

3.642379* | 10% | 2.45 | 3.52 | Evidence of Co-integration |

5% | 2.86 | 4.01 | ||

1% | 3.74 | 5.06 |

Source: By Authors, 2019. ***at 1 percent level of significance **at 5 percent level of significance *at 10 percent level of significance.

Dependent Variable: n G D P | ||||
---|---|---|---|---|

Regressors | Coefficient | Std. Error | T-Statistic | Prob. |

n I n t e r e s t _ R a t e | −0.636969 | 0.174780 | −3.644416 | 0.0010 |

n I n f l a t i o n | −0.042642 | 0.125003 | −0.341125 | 0.7355 |

n E x c h a n g e _ R a t e | −0.051043 | 0.072522 | −0.703827 | 0.4872 |

n B r o a d _ M o n e y | −0.017029 | 0.190108 | −0.089573 | 0.9292 |

C o n s t a n t | 3.805011 | 0.738028 | 5.155648 | 0.0000 |

Source: By Authors, 2019.

Dependent Variable: n G D P | ||||
---|---|---|---|---|

Regressors | Coefficient | Std. Error | T-Statistic | Prob. |

C | 0.029856 | 0.050870 | 0.586921 | 0.5623 |

D(LINFLATION(−1)) | −0.132386 | 0.102653 | −1.289642 | 0.2085 |

D(LINTEREST_RATE(−1)) | −0.354932 | 0.304485 | −1.165680 | 0.2543 |

D(LEXCHANGE_RATE(−1)) | −0.156151 | 0.091231 | −1.711601 | 0.0989 |

D(LBROAD_MONEY(−1)) | −0.553503 | 0.534754 | −1.035061 | 0.3102 |

ECM(−1) | −0.906475 | 0.214666 | −4.222732 | 0.0003 |

Source: Eviews 9 software generated results.

^{st} difference. Results of the ADF displayed GDP and Inflation (consumer price index) are integrated in order (0). Broad Money, Exchange rate and Interest rate are integrated in order (I). As a result of the different integration order of the variables, the ARDL method was preferred to be the best method for the study.

The null hypothesis of the long run relationship could not be rejected because of the decision criteria which states that, the null hypothesis can be rejected when the F-statistic is lower than the lower bound. Evidence from table three indicates that the F-statistic which has it value as 3.642379 is greater than the upper bound value of 3.52 at 10% significant level. Based on this evidence, the null hypothesis of no co-integration is rejected. These results display that there is robust relationship amongst the variables such as GPD, Inflation, Broad Money, Exchange rate and Interest rate.

The results in

In the long-run, it was revealed from

The outcomes of the short run estimates of this study are not that different from the long run estimates in exception of one variable. Thus, the exchange rate portrayed a significant influence on the economic growth but in negative direction. This implies that a unit increase in foreign exchange rate will cause a decrease in the economic growth of the country and in the inverse is also true, ceteris paribus. The short run effect of inflation on economic growth is inverse and insignificant as reveled in the long run analysis. By implication, a unit increase in inflation will cause a decrease in the economic growth of the country. Theoretically, Awan and Asalam [

From

The study investigated the game of monetary policy, inflation, and economic growth in Ghana. The study used secondary data from Bank of Ghana annual reports and International Monetary Fund (IMF) database to assess the impact of monetary policy which was proxied as Broad Money, Exchange rate, interest rate and Inflation. It was essential to perform unit root test and it was evidence that all variables estimated were stationary at first difference in exception of GDP and inflation which made the model a preferred method to be adopted for the study as there existed a relationship between the explained and explanatory variables. It revealed that inflation has negative impact on economic growth with interest rate having a significant influence in the long run whereas inflation does not influence economic growth in the long run. The study found out that exchange rate has a negative impact on economic growth of the Ghanaian economy in the long run. Finally, broad money as a proxy for money supply also shows an insignificant negative impact on economic growth. The finding is in line with (Adusei, [

Test Statistics | F-Statistics | Prob. Value |
---|---|---|

Breusch-Godfrey Serial Correlation LM Test: | 0.671083 | 0.4196 |

Jarque-Bera Normality | 1.67337 | 0.4331 |

Heteroskedasticity Test: ARCH | 0.080221 | 0.7789 |

Source: Eviews 9 software generated results.

that could help mitigate the interest rate. The Bank of Ghana can control the interest rate and true the proper medium such as the open market, thus, the purchasing of government securities. Banks will raise their prices of the securities on the open market that will result in the reduction of their rates and will then have an influence on the interest rate at large. Also, the government should minimize its interference into the dealings of Bank of Ghana and allow 90% autonomy to the Bank of Ghana.

The authors declare no conflicts of interest regarding the publication of this paper.

Wauk, G. and Adjorlolo, G. (2019) The Game of Monetary Policy, Inflation and Economic Growth. Open Journal of Social Sciences, 7, 255-271. https://doi.org/10.4236/jss.2019.73022